Lisocabtagene maraleucel in follicular lymphoma: the phase 2 TRANSCEND FL study

Follicular lymphoma (FL) is the most common subtype of indolent non-Hodgkin lymphoma, accounting for 12–32% of NHLs in North America, Western Europe and Japan. First-line treatment typically includes immunochemotherapy such as R-CHOP or bendamustine plus anti-CD20, and rituximab has improved overall survival, with 10-year OS around 80%. Nonetheless, lymphoma remains the primary cause of death for about 10% of patients, and those with progression of disease within 24 months (POD24) after first-line therapy have inferior outcomes (5-year OS of 64%). In second-line settings, various systemic options may be considered, and autologous stem cell transplantation may be used in select cases. However, patients with high-risk features such as POD24 or double-refractory disease have poor outcomes and no established standard of care, with responses diminishing across successive lines of therapy. CAR T cell therapies have shown efficacy in R/R FL in third-line or later settings, but optimal timing, especially in high-risk second-line disease, is unclear. Lisocabtagene maraleucel (liso-cel), a CD19-directed 4-1BB CAR T cell product, was evaluated in the phase 2 TRANSCEND FL study to address this gap, including a cohort of high-risk 2L patients.

Previous phase 2 trials in R/R FL after at least two prior therapies have demonstrated high activity of cellular and immune therapies: axi-cel (ZUMA-5) and tisagenlecleucel (ELARA) reported ORR 80–94% with CR rates 60–79%, and the CD20×CD3 bispecific antibody mosunetuzumab (GO29781) showed similar ranges. Prior liso-cel studies in large B cell lymphoma demonstrated deep, durable responses in 2L and 3L+ settings. Indirect comparisons suggest CAR T therapies (including liso-cel) may have advantages over mosunetuzumab in ORR, CR and PFS in 3L+ FL, though randomized data are lacking. Background data also underscore worse outcomes for patients with POD24 and in double-refractory disease, and the absence of a uniform standard of care in later lines.

Design: TRANSCEND FL (NCT04245839) is a global, phase 2, open-label, single-arm, multicohort, multicenter study evaluating efficacy and safety of lisocabtagene maraleucel (liso-cel) in adults with relapsed/refractory indolent NHL. This report focuses on FL cohorts, including high-risk second-line (2L) and third-line or later (3L+) groups. Enrollment occurred at 31 sites across North America, Europe and Japan. Period: 14 July 2020 to 27 January 2023. Participants: ≥18 years with histologically confirmed FL within 6 months of screening; ≥1 prior line of combination systemic therapy including an anti-CD20 antibody and alkylator. Cohorts: 3L+ FL included patients with ≥2 prior lines; 2L FL had exactly one prior line and required POD24 from diagnosis treated with anti-CD20 + alkylator within 6 months of initial diagnosis and/or at least one modified GELF (mGELF) high tumor burden criterion (symptoms attributable to FL, threatened end-organ function, cytopenia secondary to lymphoma, bulky disease, splenomegaly, or steady progression ≥6 months). Key high-risk features captured included POD24, double-refractory status (anti-CD20 and alkylator or anti-CD20 maintenance), and FLIPI risk. Intervention: Eligible patients underwent leukapheresis for autologous liso-cel manufacturing. Lymphodepleting chemotherapy (fludarabine 30 mg/m²/day and cyclophosphamide 300 mg/m²/day for 3 days) was followed 2–7 days later by a single liso-cel infusion at a total target dose of 100 × 10^6 CAR+ T cells (defined CD8+ and CD4+ components). Optional bridging therapy during manufacturing was permitted; PET/CT-positive measurable disease had to be reconfirmed post-bridging prior to LDC and infusion. Outpatient infusion/monitoring was allowed per investigator discretion. Assessments: Efficacy imaging by PET/CT at screening, day 29, day 90, and months 6, 9, 12, 18, 24, 36, 48 and 60. Primary endpoint: overall response rate (ORR) by independent review committee (IRC) per Lugano 2014 criteria. Key secondary endpoint: complete response (CR) rate. Other secondary endpoints: duration of response (DOR), DOR among patients with best overall response of CR, progression-free survival (PFS), overall survival (OS), safety, cellular kinetics, and patient-reported outcomes (EORTC QLQ-C30, FACT-LymS, EQ-5D-5L). Exploratory: efficacy in subgroups (≥5 patients), B cell aplasia. Safety: Treatment-emergent adverse events (TEAEs) captured from infusion through day 90. AEs of special interest included infusion reactions, cytokine release syndrome (CRS; graded per Lee et al.), neurological events (NEs; investigator-identified, graded per NCI CTCAE v5.0), MAS/HLH, tumor lysis syndrome (Cairo-Bishop), grade ≥3 infections, prolonged cytopenias (grade ≥3 at day 29), hypogammaglobulinemia, and second primary malignancies. B cell aplasia defined as CD19+ B cells <3% of peripheral blood lymphocytes by flow cytometry. Analysis sets: Liso-cel-treated set (received infusion) for safety; Efficacy set included liso-cel-treated patients with IRC-confirmed PET/CT-positive disease before infusion and a repeat baseline after any bridging therapy. ITT (leukapheresed) sensitivity analyses were conducted. Cellular kinetics assessed by qPCR in patients with available samples. PRO analysis set included patients with baseline and ≥1 post-baseline PRO assessment. Statistics: Hierarchical one-sided testing at α=0.025 controlled type I error across lines (4L+, 3L+, 2L) and endpoints (ORR, CR). Planned iNHL sample ensured ~110 FL patients treated. Exact binomial tests used for ORR and CR thresholds: 3L+ nulls ≤60% (ORR) and ≤30% (CR); 2L nulls ≤50% (ORR) and ≤19% (CR). Time-to-event endpoints summarized by Kaplan-Meier medians and 95% CIs; censoring rules specified for DOR, PFS, and OS. Treatment logistics: Median time from leukapheresis to liso-cel availability was 29 days (IQR 25–31), and to infusion was 49 days (IQR 41–55). Median on-study follow-up at data cutoff was 18.9 months (range 0.3–28.2).

Enrollment and treatment: 139 FL patients were leukapheresed (2L+). Liso-cel was manufactured for 133 (96%); 130 received liso-cel (2L n=23; 3L+ n=107). Efficacy evaluable per IRC: 124 (2L n=23; 3L+ n=101). Median follow-up 18.9 months. Baseline highlights (liso-cel-treated, n=130): Median age 60 years; stage III/IV disease 86%; FLIPI high risk 53%; POD24 from diagnosis 45%; mGELF criteria 56%; double-refractory 62%. Bridging therapy used in 38%. Efficacy (IRC; efficacy set):

• 3L+ FL (n=101): ORR 97% (95% CI 91.6–99.4; P<0.0001); CR 94% (95% CI 87.5–97.8; P<0.0001). Median time to first response 1 month (0.6–3.3). DOR NR (95% CI 18.0–NR) with 12-month DOR 82% (95% CI 72.5–88.4). PFS NR (95% CI 19.0–NR) with 12-month PFS 81% (95% CI 71.4–87.2). OS NR with 12-month OS 92% (95% CI 84.8–96.0).
• 2L FL high-risk (n=23): ORR 96% (95% CI 78.1–99.9; P<0.0001), all responders CR (CR rate 96%; 95% CI 78.1–99.9). Median time to first response 1 month (0.8–2.8). DOR NR (95% CI 19.3–NR) with 12-month DOR 90% (95% CI 64.8–97.4). PFS NR (95% CI 20.2–NR) with 12-month PFS 91% (95% CI 69.5–97.8). OS NR with 12-month OS 96% (95% CI 72.9–99.4).
• ITT sensitivity: Similar response rates (3L+ ORR 93%, CR 90; 2L ORR 92% with all CR).
• Subgroups: High ORR/CR and 12-month DOR and PFS across subgroups including POD24, double-refractory, high FLIPI, high tumor burden (mGELF), and with/without bridging therapy. IRC vs investigator response assessments showed 95% concordance. Safety (liso-cel-treated, n=130):
• Grade ≥3 TEAEs in 75%; serious TEAEs in 25%. Most common grade ≥3: neutropenia 58%, thrombocytopenia 10%, anemia 10%; febrile neutropenia 6%.
• CRS in 58% (grade ≥3 in 1%); median onset 6 days; median duration 3 days; managed with tocilizumab and/or steroids in 25%. No grade 4/5 CRS.
• Neurological events in 15% (grade ≥3 in 2%); median onset 8.5 days; median duration 3.5 days; managed primarily with corticosteroids; no grade 4/5 NEs.
• Grade ≥3 infections: 5% within 90 days; late grade ≥3 infections in 3 patients (>90 days). Prolonged cytopenias at day 29 in 22% (neutropenia 15%, anemia 5%, thrombocytopenia 15%) with most recovering to ≤grade 2 by day 90 (neutropenia 90%, anemia 83%, thrombocytopenia 58%). Hypogammaglobulinemia AEs 5%.
• MAS/HLH in 1 patient (1%; grade 5; in 2L cohort). Second primary malignancies in 3%.
• Deaths on-study: 13 total (1 pre-infusion; 12 post-infusion, including 4 due to disease progression). Two deaths considered related to liso-cel (one MAS/HLH; one PML after 90-day TEAE period). Cellular kinetics and B cell aplasia: Median tmax 10 days; median Cmax 42,026 copies/µg; AUC(0–28d) 260,274 days×copies/µg. Persistence detected in 41% at 18 months. B cell aplasia increased from 77% at baseline to 99% post-infusion, remained ≥91% through day 90, then declined (57% at month 18). Patient-reported outcomes: High completion rates (70–95%). Improvements from baseline by day 29 in key domains (fatigue, pain, global health status, FACT-LymS) generally maintained through month 18; several domains showed statistically significant and clinically meaningful improvements over time.

The study addresses the unmet need for effective therapies in relapsed/refractory FL, particularly among high-risk patients (POD24, double-refractory) lacking a standard of care and with diminishing responses after successive lines. Liso-cel achieved very high ORR and CR rates in both 3L+ and high-risk 2L cohorts, with rapid responses and durable disease control at approximately 17–19 months median follow-up. Efficacy was consistent across key poor-prognosis subgroups and irrespective of bridging therapy, suggesting broad activity. Safety was manageable, with low rates of severe CRS (1%) and NEs (2%), and low utilization of tocilizumab/steroids compared with some other therapies. Comparisons with published single-arm studies of other CAR T products and bispecific antibodies suggest liso-cel’s efficacy and tolerability are favorable in 3L+ FL, though direct cross-trial comparisons are limited by design differences. Together, these findings support earlier use of CAR T therapy with liso-cel, including in high-risk 2L patients, as a strategy to achieve deep remissions with acceptable toxicity.

In a large phase 2 experience of R/R FL, including the first reported outcomes of CAR T therapy in high-risk second-line FL, liso-cel produced high response rates with a predominance of complete responses and durable disease control, alongside a favorable safety profile marked by low rates of severe CRS and neurological events. These results support liso-cel as a potential treatment option across 2L and later-line R/R FL, including patients with high-risk features such as POD24 and double-refractory disease. Longer follow-up and comparative or real-world studies will further define durability, optimal timing, and patient selection.

• Single-arm, open-label design without a comparator limits causal inference and cross-study comparisons.
• Median follow-up ~19 months; many patients were censored with ongoing response, so longer follow-up is needed to mature DOR and PFS estimates.
• The 2L cohort size was relatively small compared with 3L+, limiting precision of safety and efficacy estimates and subgroup comparisons.
• Outpatient monitoring numbers were limited by regional regulations and the COVID-19 pandemic.
• Study population skewed younger and included a high prevalence of high-risk features; generalizability to broader FL populations requires caution.